• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.1
• /
• pp.20-28
• /
• 1992

Co-articulation is one of major factors that make connected word recognition difficult. This Study Considers the fact that the head Part Of the following word is changed by the Preceding word in a connection point, by applying the co-articulation model, and adj usting the following word .We choose a critical damping second order linear system for the co-articulation model, combining a one-stage DP matching recognition algorithm with this model, and Investigating the effects. The recognition experiment is carried out for 35 Korean 4 connected digits spoken by 5 male speakers, and recognition rate Is upgraded by 4.7 percent.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.4
• /
• pp.355-360
• /
• 2015

In this paper, we propose the design of Radial Basis Function Neural Network based on PCA in order to recognize handwritten digits. The proposed pattern classifier consists of the preprocessing step of PCA and the pattern classification step of pRBFNNs. In the preprocessing step, Feature data is obtained through preprocessing step of PCA for minimizing the information loss of given data and then this data is used as input data to pRBFNNs. The hidden layer of the proposed classifier is built up by Fuzzy C-Means(FCM) clustering algorithm and the connection weights are defined as linear polynomial function. In the output layer, polynomial parameters are obtained by using Least Square Estimation (LSE). MNIST database known as one of the benchmark handwritten dataset is applied for the performance evaluation of the proposed classifier. The experimental results of the proposed system are compared with other existing classifiers.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.175-182
• /
• 2023

In this paper, we propose a study compared and analyzed the CPR results and muscle fatigue of the three groups. There is a group that counts loudly when compressing the mannequin's chest (Group A), a group that breathes autonomously without counting (Group B), and a group that breathes abdominally without counting (Group C). Twelve people were assigned to each group, and after performing chest compressions for 5 minutes, the results of CPR were analyzed using a program connected to the mannequin, and the muscle fatigue of the performers was analyzed using wireless electromyography. The most efficient method was found to be group B. If we only look at the speed and depth of compression within the normal range, Group C would be more efficient, but Group B showed significantly lower muscle fatigue, and Group A did not reach the normal range in depth of chest compression and muscle fatigue was the highest. Group B was also found to be the most accurate in hand positioning accuracy, and was also found to be the most efficient in maintaining concentration on chest compressions.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.8
• /
• pp.96-105
• /
• 1998

The aim of this paper is to propose the method of WDHMM(Weighted DHMM), using the MLP-VQ for the improvement of speaker-independent connect-digit recognition system. MLP neural-network output distribution shows a probability distribution that presents the degree of similarity between each pattern by the non-linear mapping among the input patterns and learning patterns. MLP-VQ is proposed in this paper. It generates codewords by using the output node index which can reach the highest level within MLP neural-network output distribution. Different from the old VQ, the true characteristics of this new MLP-VQ lie in that the degree of similarity between present input patterns and each learned class pattern could be reflected for the recognition model. WDHMM is also proposed. It can use the MLP neural-network output distribution as the way of weighing the symbol generation probability of DHMMs. This newly-suggested method could shorten the time of HMM parameter estimation and recognition. The reason is that it is not necessary to regard symbol generation probability as multi-dimensional normal distribution, as opposed to the old SCHMM. This could also improve the recognition ability by 14.7% higher than DHMM, owing to the increase of small caculation amount. Because it can reflect phone class relations to the recognition model. The result of my research shows that speaker-independent connected-digit recognition, using MLP-VQ and WDHMM, is 84.22%.

• 한국정보교육학회:학술대회논문집
• /
• 2007.08a
• /
• pp.273-279
• /
• 2007

수학은 생활의 기본이 되는 도구 교과이다. 그 중에서 아동들이 가장 중요하지만 어려워하고 후속 학습에까지 영향을 미치는 것은 구구단이다. 고학년이 되어서까지 곱셈 구구를 못 외우는 학생들은 지도가 어려울 뿐만 아니라 이후에도 지속적으로 수학 학습에서 뒤떨어지게 된다. 하지만 암기가 중심이 되는 부분이라 다른 것들보다 학습의 흥미가 떨어지고 시기를 놓치면 학습이 점점 더 어려워지는 단점이 있다. 이에 지금까지 어려웠던 어렵고 암기위주로 흘러가던 구구단을 음악과 연결시켜 보는 것은 어떨까란 생각을 하게 되었다. 여기에서 쓰이는 음악들은 모두 초등학교 교과서에 있는 음악으로 아동들이 친숙하게 느낄 수 있는 음악들이다. 이 소프트웨어는 2학년 뿐만 아니라 구구단을 어렵게 느끼는 고학년의 학습에도 도움이 되도록 제작하였다. 프로그램은 전체 구구단인 2$\sim$9단을 다루도록 하였으며, 각 단은 4개의 파트로 구성되어 있다. 1단계에서는 1-각 수까지 반복해서 세는 연습을 하였고, 2단계에서는 반복했던 수를 연결하여 띄어세기 연습, 3단계에서는 2단계에서 했던 띄어세기를 확장하여 구구단 연습을 해 볼 수 있도록 하였다. 정리단계에서는 각단의 구구단을 랜덤으로 배치하여 구구단의 복습이 될 수 있도록 하였다. 2,4,5,6,8단은 각기 박자에 맞춰서 구구단을 외우는 것이고 5,7 단은 박자에 맞추기 어려우므로 글자수에 맞춰서 구구단을 외우도록 하였다. 박자에 맞추는 것이 주가 아니라 수학을 배우는 것이 중심이므로 틀린 박자에 맞는 숫자를 누르는 것은 반응하지 않도록 하였다. 구구단 중간에 있는 수까지를 전부 제시함으로써 큰수와 띄어세기 그리고 구구단과의 관계를 이해할 수 있도록 하였다.

• The Journal of Information Technology
• /
• v.4 no.1
• /
• pp.71-79
• /
• 2001

This paper presents an adaptive back propagation algorithm that update the learning parameter by the generated error, adaptively and configuration of the range for the initial connecting weight according to the different maximum target value from minimum target value. This algorithm is expected to escaping from the local minimum and make the best environment for the convergence. On the simulation tested this algorithm on three learning pattern. The first was 3-parity problem learning, the second was $7{\times}5$ dot alphabetic font learning and the third was handwritten primitive strokes learning. In three examples, the probability of becoming trapped in local minimum was reduce. Furthermore, in the alphabetic font and handwritten primitive strokes learning, the neural network enhanced to loaming efficient about 27%~57.2% for the standard back propagation(SBP).

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.4
• /
• pp.19-27
• /
• 2002

We proposed an auto-tuning method of learning rate for performance improvement of backpropagation algorithm. Proposed method is used a fuzzy logic system for automatic tuning of learning rate. Instead of choosing a fixed learning rate, the fuzzy logic system is used to dynamically adjust learning rate. The inputs of fuzzy logic system are ${\Delta}$ and $\bar{{\Delta}}$, and the output is the learning rate. In order to verify the effectiveness of the proposed method, we performed simulations on a N-parity problem, function approximation, and Arabic numerals classification. The results show that the proposed method has considerably improved the performance compared to the backpropagation, the backpropagation with momentum, and the Jacobs' delta-bar-delta.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.12
• /
• pp.21-30
• /
• 1993

In this paper, we propose a connected Korean digit recohnition system employing neural networks and lexical constraints of the Korean digits. In the proposed recognition system, firstly, each frame of digit string is labelled by phoneme classification neural networks.which are trained with the reference phoneme segments extracted form an isolated digit based on the position information. And, the frame labels are combined with each other for constructing the phoneme segments. Then, these segments are combined to form a digit candidate using the digit combination rules. The digit candidate is decided based on the condition for digit decision. If the condition is not satisfied, the digit candidate is further recognized using the digit decision neural network in the next step. In our approach, the neural networks are trained with 10 isolated digits uttered by 5 male speakers. To investigate the performance of the proposed recognition system, an intensive computer simulation on the 30 connected digit strings uttered by 5 male speakers is performed. The simulation result indicates that 95.6% digit recognition rate and 82% digit string recognition rate are provided by the proposed Korean digit recognition system.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.71-76
• /
• 1999

본 논문에서 제안한 예측형 RBFN(Radial Basis Function Network)은 HMM과 신경망을 결합한 하이브리드 구조이다. 이 신경망은 HMM으로 추정한 확률분포 파라미터를 사용하여 중간층의 활성화 함수의 출력을 결정하고, 중간층과 출력층의 연결강도만 네트워크 내에서 학습한다. 그리고 HMM으로 추정한 확률분포 파라미터는 두 가지 방법으로 예측형 RBFN에 이용하였다. 첫 번째는 HMM의 각 상태의 혼합수 만큼의 중간층 유니트를 주는 방법이고, 두 번째는 HMM의 혼합수$\times$출력분포수 만큼의 중간층 유니트를 주는 방법이다. 실험결과, 예측형 RBFN은 다른 방법들의 결과보다 $4.5\~6.5\%$ 저하된 결과를 보였지만 다른 신경망에 비해서 학습 반복 횟수를 작게할 수 있었으며 전체 학습시간을 대폭 단축할 수 있었다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06c
• /
• pp.287-290
• /
• 1994

음성 인식 시스템의 개발을 위해서는 음성 데이터베이스구축이 중요한 과제의 하나로서, 많은 시간과 노력이 요구된다. 개별적인 음성데이타베이스 구축에 따른 중복 투자를 줄이고 다양한 인식 알고리듬의 성능 비교와 국내 음성 인식 기술의 발전을 위해서는 벤치마크 시험을 위한 공통의 음성 데이터베이스가 필수적이다. 본 논문에서는 한국과학기술원 통신연구실에서 제작한 한국어 음성 데이터베이스에 관하여 기술한다. KAIST 음성데이타베이스는 자동통ㅇ역을 N이한 무역 상담과 관련되 3,000 단어 규모의 연속어를 비롯하여, 가변 길이 연결 숫자음, phoneme-balanced 75 고립단어, 지역명 관련 500 고립단어, 한국어 아-세트로 구성되어 있다. 이 음성 데이터베이스의 구축을 위하여 사용된 태스크선정 절차, 녹음 방법, 규격, 및 기대효과 등 세부사항을 기술한다.